SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.01704 · 2024

Weight Clipping for Deep Continual and Reinforcement Learning

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Many failures in deep continual and reinforcement learning are associated with increasing magnitudes of the weights, making them hard to change and potentially causing overfitting. While many methods address these learning failures, they often change the optimizer or the architecture, a complexity that hinders widespread adoption in various systems. In this paper, we focus on learning failures that are associated with increasing weight norm and we propose a simple technique that can be easily added on top of existing learning systems: clipping neural network weights to limit them to a specific range. We study the effectiveness of weight clipping in a series of supervised and reinforcement learning experiments. Our empirical results highlight the benefits of weight clipping for generalization, addressing loss of plasticity and policy collapse, and facilitating learning with a large replay ratio.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2407.01704")
get_code_for_paper("2407.01704")
have("2407.01704")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2407.01704.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2407.01704.svg)](https://syntology.ai/paper/2407.01704)

Connect an agent — have() is free.